The build is the easy part.

Identify the problem first

Amanda’s first takeaway from John Beyer’s conference-room tool: the build only worked because he named the actual failure first, IT never learning what a booked room needed, rather than starting from the technology. John’s answer when asked what language he coded in stuck with her too: English is the programming language I know.

Build it alongside IT, not around it

A security-conscious IT team is not automatically a fan of Claude, since it needs broad permissions and effectively runs a virtual machine on the user’s own computer. Amanda’s standing move: ask Claude to draft exactly what to tell IT, security concerns included, before you roll anything out, or you risk building something IT simply will not let you ship.

Who owns this now?

Debbie’s question the panel never put to John directly: what happens to what he built if he leaves, or wins the lottery and walks. Amanda takes it further, John built most of it on his own personal time, and no firm has really had to answer who owns a tool an administrator builds for it. Tech companies write IP agreements for exactly this. Law firms mostly have not had to yet.

Not booths. Neighborhoods.

Debbie’s report from a 5,600-person ILTA: vendor booths have grown into neighborhoods complete with living rooms, baristas, and their own conference rooms, with parties headlined by Lady A, Cheryl Crow, and Little Big Town. The number counts every exhibitor and their staff, not just firm attendees, but 248 sponsors still means any firm can walk that expo hall and find a real use case.

The real question is not how a firm adopts AI

That framing is too big to answer, Debbie argues. The useful version is narrower: name three problems the firm has had for years that an AI tool could actually solve, and check what the firm already owns before buying something new. Her live example, a prospect shopping for new tools who did not realize the Centerbase she already owned could do three of them.

Tech spend is headed from 5 percent of revenue to 7

Firms have historically spent 3 to 5 percent of revenue on technology. Amanda expects that to move to 5 to 7 percent within five years, with 5 percent marking the firms falling behind and 7 the firms staying ahead. On a mid-sized firm’s budget, that swing alone can mean hundreds of thousands of dollars that has to come from somewhere, and rates are already climbing to cover it.

The rate is a phantom ceiling

Debbie’s math: raise your rate from $500 to $600 an hour to pay for an AI tool, and a job that used to take an hour now takes half that. The client pays $300 instead of $500, and the firm keeps more, but almost nobody actually raises the rate that far, because it feels like hitting an invisible cap. Rob pushes back that he would rather pay more for the faster, better outcome, and Amanda counters that in practice, contingencies (a will that turns into an estate plan with three kids and an ex-marriage) make flat-rate math far harder than the thought experiment.

Boxed in by the bar, breaking the ceiling anyway

Bar rules keep law firms locked into pricing by the hour worked, unlike Debbie’s own consulting business, which can put a client on a flat monthly retainer with no one telling her how to price it. Debbie’s counterexample that a ceiling can move anyway: a New York firm stuck at $995 an hour went to $1,050 for January after she showed them competitors already charging $1,200 to $1,400. Nobody pushed back.

Transcript

Finally, actually live

Amanda: Yeah.

Rob: No, I think we’re live. We are live this time. Actually, I can tell that we’re live this time. It’s been a few weeks

Debbie: It’s good.

Rob: Since we’ve been live. We’ve been having a lot of technical difficulties with the platform we use, which I’m not going to name. but last

Amanda: Ha ha.

What did you learn from John last week?

Rob: Week we did record a session. It was an amazing session. I watched it back since I was out. Two things I’d love to start with. You know, Debbie, I’d love to talk about ILTA, but before we get there, Amanda, what did you learn last week with John on? Like what were your takeaways?

Identify the problem first

Amanda: Well, first of all, he’s really far ahead and advanced of most people who are running law firms right now. And it’s always kind of hard to see that and think I’m so far behind, because I’m nowhere near where he is, right? Before the call he was showing me this second brain he has invented and all these different things he’s working on, and I’m just like, my God, I’m so far behind him. It’s very hard to think about that. But what I really appreciated is that the first project he rolled out from his firm is solving a problem that so many of us have, and it started with what the famous Debbie Foster always says, which is identify your problems first. The problem he identified was that when people were booking conference rooms in his offices, they didn’t know what was in each room, and then IT wasn’t getting notified that the request was coming through, and calls weren’t getting set up, and the right drinks weren’t being set up. We all have these problems. We use forms you have to submit with the thing. Other large firms use software that is designed to manage spaces, but they’re not a large firm, they’re a mid-sized firm. And so he built a solution in Claude. I thought it was really fascinating because in concept it’s very simple, and I think the best thing I learned from him is when he’s asked what programming language do you know, his response is, English is the programming language I know.

Debbie: Right.

Rob: I remember hearing that and I was like,

Amanda: And

Rob: Okay, I guess. Yeah, makes sense.

Amanda: I mean, Rob, do you know Python or Java or

Rob: Well technically those are still the programming languages. You’re just the interface and how you do it is different. Yeah. One hundred percent.

Build it alongside IT, not around it

Amanda: But you don’t have to know then. Right, right. So I think that was a big takeaway for me, and also just that he continues to build every day, and that’s where he’s been able to add value. The second thing is we’ve talked a lot about the handoff and the handshake between administrative operations and the IT department. And he talked a lot about, Claude in particular is not always a favorite of your IT team, specifically your security team, because it needs a lot of permissions. It basically runs a virtual machine on your machine, and if you have security-conscious IT people, they might not be comfortable with you giving all those permissions. So his advice was something I always do, which is if Claude is spitting out something and it starts to sound technical, I will always say, I need you to draft exactly what you want for the IT team. Be sure you address security concerns they might bring up, in addition to what settings you might have. So it starts that conversation, and it’s having it for you. But it was a good point, because we can do all the things we want and build everything we want, but if IT comes back and says you can’t roll that out here, then you’ve got to be building it alongside them.

Rob: I think really good point. I would say a few things. The most important part when you’re coding with Claude, like he’s doing, is the planning. Spending a lot of time on the planning, the structure, the architecture of the documents that Claude’s gonna use to build it. And Amanda, to your point, you can actually build that in. Like, hey, every time you propose something, make sure you explain it this way. Very similar to memory, but you can actually put that into your project so it doesn’t go wild or anything like that as it’s building.

What happens if John leaves?

Debbie: So one other thing that I we didn’t actually address this with John on the call, but if I was running a law firm, the first thing that I would think was, what happens if John leaves?

Rob: Mm-hmm.

Debbie: So it’s all fine and well and cool to build stuff. But how do you document what you built? How do you bring other people into the fold in case forget about if John leaves, what if John wins the lottery and he’s like, gotta go? You know? how do firms prepare for that? And that might be more than we can get into in this conversation, but I think it applies beyond just build something with Claude, right? You’re we actually have a client and Harvey, I shouldn’t have said the name. Dang it, flip me out.

Amanda: It’s okay. We’re

Rob: That’s okay.

Amanda: Saying Claude’s name, it’s okay.

Debbie: Hired one of our clients, people who had built out some really amazing playbook stuff. And the truth of the matter is, some of these software companies, I’m not saying that Harvey did this, by the way, should have never said the name. can pay

Amanda: Sure.

Debbie: A lot more than a law firm is paying for those rules. And so I think that this is a this is a debate. This is John Beyer built something super freaking cool and I have ten more questions.

Rob: I love it. I mean we’ve we you know with Claude, especially when you’re building within an organization, for an organization, there’s a difference between building something to an MVP and productionalizing something, which is what you’re talking about. who’s gonna maintain it, who’s gonna handle the bugs? The building’s the fun part. Then you have to think about well, what does that roadmap look like? And Debbie to your point. Maybe that’s a consulting offering, right? Or as you’re looking, we’re about to enter budgeting season, do you bring somebody in that can help support from an IT infrastructure that’s maybe more of a, you know, a different role than an IT and security role? I don’t know.

Who owns this now?

Amanda: I my head immediately went to John told us and his firm’s probably gonna murder me, but John told us he built a lot of this in his personal time. Who owns this now? And I mean that’s a that’s a big, big step that a lot of companies deal with on a regular basis, like tech companies. But this is the first time a law firm has to think who owns something if an administrator built it? And they

Debbie: Yes.

Rob: In tech companies generally have IP agreements with anybody building, right? Hmm,

Amanda: Yeah, right, right, right.

Rob: That’s tough. Well Go back. If you’re interested

Amanda: Hehehehe

Over to ILTA

Rob: In learning more, go back on our website. We have the episode. It’s live up there. So go sidebar.ai and check it out. It’s worth the 30 minutes. we also keep the cliff notes below in the episode on the website. So if you just want to scroll through and find stuff, it’s easy to do. Now, Debbie, I addressed that question to Amanda because I was Playing around in Mexico and forgot that you were not only at ILTA, but you showed up as well. But I want to flip over to ILTA for a minute. ILTA was last week, biggest conference I would say in legal. 5,600

Amanda: Fifty six hundred attendees they had this year.

Rob: Attendees. So huge, bigger, twice as big as a lot of the other larger conferences, at least, that I’ve attended. What are your what are your takeaways? What’s the market doing?

Debbie: The market is spending money.

Amanda: Tell me about it. Jeez.

Booths the size of neighborhoods

Debbie: Right. So I had told my team, I’m like, you guys should see these booths and they’re like, Are they like t huge? And I was like, It’s not, like I can’t so I walked around when there was no one in the expo hall and I videoed like walking in the front door, seeing the big giant booth that was right in front of you. But then I stopped calling them booths. I said they’re neighborhoods. So you want to go to like the Harvey neighborhood or to the Litera neighborhood or to the Clio neighborhood. I mean, these booths had bookshelves with knick-knacks. They had like living rooms. They had they had they had like one of had a big cowprint rug under a under a coffee table. Like they had conference rooms built into the booth, so you could just be like, Let me take you into my conference room. I it was they all had baristas. That were but you they had their logos in the foam on the coffee. Like I also the parties that they threw headlined by Lady A, Cheryl Crow, Little Big Town,

Rob: Wow.

Debbie: Were three of the headliners at the parties that these companies were throwing. I I’ve never seen anything like it. In I’ve never seen anything even kind of like it. And When I said that to someone last week, they said even at legal week. And I said, even at legal week, like it’s it was unprecedented. but there were 5600 people there. Now, to be very clear, when we report conference numbers, we are including the exhibitors and all of their people. So some

Amanda: Right.

5,600 people, and what that number actually counts

Debbie: Of these exhibitors brought 50 was a number that was thrown out. 85 was a number that was thrown out. 125 was a number that was thrown out. People to come attend the sessions, speak at the sessions, man the booth, go to the social events. In addition to the jam-packed agenda that ILTA put on, many of the larger sponsors also had their own sessions in their own session rooms where they’re talking about products. Roadmaps, entertaining their clients. I went to a few of the company updates. Mind-blowing what they’re showing. Now, granted, a lot of what they’re showing is vaporware, roadmap stuff. Like it’s not, here’s what’s coming, here’s what we’re doing. That’s their company update, but it’s really impressive stuff. I mean, I continue every year, I wonder how ILTA will do it bigger and bolder and better. And every year it’s bigger and bolder and better. And I’ve I don’t know, like I don’t know if they’re gonna like somebody’s gonna bring Elvis and Dolly back from you know like for next year because I feel like that’s about also Dolly died while we were at ILTA in Nashville.

Rob: Ooh.

Amanda: And were

Debbie: And

Amanda: You at the Gaylord? Were you guys at the Gaylord? Yeah.

Rob: So when I hear that, Debbie, I’m laughing, you’re saying all these things, but then I’m thinking, I’m somebody from a law firm that was sent there to come back with something, right? And your firm doesn’t care about how many drinks you had or how many parties you went to. They sent you there for a reason. So if somebody were to ask you, what is the state of the market, what’s happening right now, what would you say based on ILTA?

The real question is not how a firm adopts AI

Debbie: So I would say it’s a lot of what we’ve been talking about. There’s a lot of noise and there’s not a lot of actual implementation. Everyone’s struggling with the same things. Like you, I think that going back to the example with John, John found a problem and he took a thing and he built a thing to solve that problem. And so there’s one school of thought where you’re thinking like, how will a law firm adopt AI? That question is so big. And I think a lot of people come to ILTA to try to under came this year to understand like how will my law firm adopt AI? But the real answer is what three problems that you have and have had for a long time will be solved with an AI tool? and you would be you would be missing lots of opportunity if you didn’t also think. What about all the stuff I already own? How could I solve some of those big problems with what I already own? Because Rob, I I’m not gonna say any names, but this is a Centerbase thing.

Amanda: Yeah.

She already owns the tool that does this

Debbie: I was on a call with a prospect for Affinity yesterday who wanted to buy some new things. She already owns Centerbase. And I was like, So I just want to point out, like. Three of those things, Centerbase can already do them. And she was like, I don’t think it can. And I said, Can you share your screen? And she

Amanda: Yeah.

Debbie: Shared her screen and I said, open that, click on that. And she said, You have got to be kidding me. I said, not. She said, I wonder if this activity tab, and she clicked on it and she goes, Yep. Can I just tell you in the demo that I was on yesterday how cool I thought it was that other program could do this? And I had no idea. That Centerbase

Rob: It

Debbie: Could do.

Train on it every quarter

Rob: Yeah, and to that point, there’s an industry stat, and I’m sure it’s changed, that you adopt twenty percent of a piece of software. That’s something you preach every day because of the business you’re in, and Amanda has done this well too, is you’ve got to consistently train. You can’t just do it once and move on to the next new product. You have to do it every quarter. And it’s becoming more and more important as the technology is rolling out faster. Debbie, I agree with you. We’ve gone as an industry from announcing releases when things are being released, to announcing releases before they’ve even been worked on. Right?

Amanda: Our builds.

Debbie: Yes, because we had like keeping up with the Joneses. It’s just a different version of keeping up with the Joneses.

Rob: Very much so. And it’s confusing. Go ahead.

Can firms actually afford this?

Amanda: Debbie, I have a question for you. You started out by saying people are spending money, and I know what you were talking about is the business partners, the vendors, are spending money because client acquisition costs right now are higher than they’ve probably ever been, because you’ve gotta break through the noise, right? Every company is doing something and you’ve got to break through, and that’s what we’re feeling on our side too. So but our firms, how are they affording this? I keep going back to that. When you’ve got two of these huge competitors that are going to single-handedly increase your tech spend by one percent of your revenue every year, that’s a lot. It’s a big percent, right? So what are firms saying, what are you hearing, you know, firm leaders, IT leaders, talking about with this?

Debbie: I think there’s a lot of conflict. because there’s the keeping up with the Joneses aspect, but there’s also the reality of using a tool and watching how it simplified something for you and saying there’s a there. But then that challenge that we run into spend more money to make less money is one assumption, right? I have to spend this money and I’m gonna make less money. But there’s also the reality of that where you say efficiencies will always level will always raise the bar, right? Well, whatever that efficiency is. My joke has been the whole like, why would I buy a tool that lets me do it in one hour when I can build for five? And I’m like, well, if you that’s the goal, just get a typewriter, throw your computer away. It’ll take you hours. Like you

Amanda: Right, right, right.

Debbie: Can find a way to do things less efficiently. So some of this is like how do we get law firms to understand that there is a cost to being more efficient? So maybe it’s a rate issue, it’s not a value billing issue, but there is also a value billing issue. So I think to come back around to your question, it’s very challenging. There’s noise, and then there are what seems to be insurmountable business challenges. Right? Touching partner comp, touching path to partnership, touching leverage. Those things that these tools absolutely have an impact on. It’s the pull the string. You touch one thing, five other things move. And I don’t I think that people are seeing things they really like and that they want. But pulling that little thread makes you have to address some of these other things that are just yeah,

Rob: The domino effect. Yeah.

Debbie: It’s just harder to talk about. So, but I saw a lot of excitement about what’s coming. So there were 248 sponsors at ILTA, so 248. any law firm could walk into that expo hall and could find a use case. For let’s just say a hundred of the two hundred and forty eight.

Amanda: So

Debbie: What do you do?

Amanda: I yeah, what do you do? What indeed? Well so I wonder, Rob, from a from a vendor standpoint, are you all like it’s there’s so much pressure. And so sometimes I worry that like vendors don’t understand like the pressure that we are under with pricing. Is some of that eventually going to are we gonna hit a bubble on some of this?

Rob: Maybe, right? There’s only so much wallet to grab wallet share of, right?

Amanda: Right.

Is there a spending benchmark?

Rob: I mean, Debbie, I think I’ve actually seen an article that you were quoted in on this, and Amanda, I’d ask you this too. When we talk about that wallet share and how much you’re spending on technology, is there a benchmark, as far as percent of your annual expenses or of your revenue, that a healthy firm should be considering? Or is that too, it depends.

Three to five percent, heading to five to seven

Amanda: No, it’s not. I mean, like historically firms have spent three to five percent. of revenue on technology. But what I am seeing and what I entirely expect in the next five years is it’s gonna go five to seven. Five percent is the behind firms and seven percent is the ahead firms. That or more, right? But that’s that two to three percent has gotta come from somewhere. And it’s rates are already having to go high to cover the cost of labor and to cover the which is very high. And so, like, those that those percentages, those two percent wiggle point percentages for a mid sized farm could be hundreds of thousands of dollars, you know.

Debbie’s rate math

Debbie: So Amanda, let’s stay on the rate thing for a second. Now I

Amanda: Okay.

Debbie: Have to get my head in my mind in math mode. Okay,

Rob: Yeah.

Debbie: Let’s just say that.

Amanda: Do I need to grab my phone and grab a calculator?

Debbie: No, I think I have easy math for all of us.

Amanda: Okay.

Debbie: Say that. Our hourly rate is five hundred dollars an hour.

Rob: I got my calculator out. Go ahead.

Debbie: And that thing that we’re about to do is going to take an hour. We’re gonna build a client five hundred dollars. Now, let’s say that we buy an AI tool. And we raise our rates to six hundred dollars an hour to pay for the tool. And now it only takes us a half an hour. And we have $300 and it only costs the client $300 instead of $500, but they paid $600 an hour instead of $500 an hour. So is our rates are already maxed out like a what do you call that? Like a cap, like a ceiling

Rob: Ceiling? Yeah.

Amanda: Cap. Yeah.

Debbie: That we feel like we can’t go above. Because if you are my client and you say, I’m only willing to pay five hundred dollars an hour, and I say fine, it’ll take me an hour. But if you’re the client and you’re like, sure, six hundred dollars, but it’s only gonna take me a half an hour, you guys make more, the client pays less, but the rate is like this phantom cap that we can’t go above. Does that make sense?

Rob: So Debbie, I’d argue,

Amanda: Yeah, except that’s

Rob pushes back: I’d pay more for the value

Rob: Hold on, from a consumer standpoint that doesn’t work for me. If somebody came to me and said, hey, I need this, and they said it’s gonna take me an hour, $500, or they said, hey, I need this, and it’s six hundred dollars, I’d actually rather have the latter, right? Even if it takes less time with the AI tool, I’d rather have the value.

Amanda: Yeah, I mean

Rob: Time with the AI tool, I rather have the

Amanda: There’s just

Rob: Latter. The value.

Amanda: It’s just so easy to say in concept. It is

Easy in concept, impossible in actuality

Rob: It is.

Amanda: Impossible in actuality. It is impossible. Even on things that you think are gonna be like, you guys, let’s talk about it from I want to do a simple estate plan. Great. It’s me and my husband and we have a house and we just need a simple estate plan. Great. Halfway through. Okay, that’s five thousand dollars. Halfway through. and by the way, I forgot to tell you. There’s this.

Debbie: We have three kids.

Amanda: Right. Or we I forgot to tell you, she has a

Rob: LLCs.

Amanda: Son from the first marriage that needs to be, you know, whatever. And so all of it has to you have to have both a flat rate and then all the contingencies that have to be hour, but you have to lay out. Like when does it click from flat rate to hourly? We’ve been talking about like probates, like do you do it as a percentage of the size of the assets or based on certain concepts? M&A deals. You could it you could do as a percentage of the fee, but then there could be five million pages of due diligence. I’m exaggerating. So it’s very easy to talk about in concept. It’s just very hard to do in actuality.

Rob: Debbie, I think you’ve done this at Affinity.

Breaking the rate ceiling in practice

Debbie: We have. I wanna just say one quick thing though. Amanda, what you’re talking about? Totally get it. The thing that I was trying to talk about with the rate is how do we remove the mindset of our rates can’t go above this, right? Like I just I worked with a firm last year, whether this is ridiculous and absurd or not, but they were at 950 an hour and they were raising their rates. New York firm. They’re raising their rates, and they were like, we can only go to 995. And I’m like, guys, we just looked at your competitors. They’re charging $1,200 and $1,300 and $1,400 an hour. Go to $1050 for January. And if people push back, you can go back to $9.95. No one pushed back. So there the proof in that was there wasn’t a thousand dollar marker. So my point in the like we are raising our rates and also. You should expect to see some efficiencies, even though our hourly rates are higher. Our bill amounts should be lower because we’re leveraging AI. Maybe not at our whole entire firm, but our real estate group is really leaning into doing work efficiently, and our traditional rate has been this much an hour. It is going up to the like, how do we, even in the hourly world, forget about having to move into flat rate billing? How do we tell the story of hourly rate efficiency and why we can justify a higher hourly rate? Food for thought.

Rob: Is there any value to speed? Right? Faster outcomes?

Boxed in by the bar rules

Amanda: Sure, but we’re still constricted by the bar rules. You’re so constricted by the bar, that’s the precedent of the legal industry. It’s what makes all of the AI adoption and the pricing around it the most difficult thing, is that there are laws restricting how we can price. Like you said, Debbie, you can put people on a retainer, and they get up to so many hours and pay this amount per month. She can do that. Law firms can’t.

Debbie: Ain’t nobody telling me what to do, right? Yeah.

Amanda: Ain’t nobody telling her how to price. Rob, same thing with you, nobody’s gonna tell you how much you can charge for software. But we can only charge how much we work, unless you have an alternate model. And if you are a firm that has historically charged by the hour, and everybody’s talking about this, this is not new, but if you’re a firm that has historically charged that way and all of a sudden has to switch, it’s hard. Now Debbie, to your answer about the rates, it’s not always that we think there must be a cap. First of all, it’s really hard to find out what your competitors are charging, by the way, just food for thought.

Debbie: Yeah.

Amanda: Unless you’re taking a senior real estate attorney and comparing it to one or two other senior attorneys, it’s very hard to figure out what your competitors are charging. It’s not like people are sending out their regular legal work to bid, most of the time.

Rob: Mm.

Amanda: So that can be very difficult. But sometimes there are other mental blocks, like a practice area, say family law, might not be able to charge enough. But they might have an M&A attorney, and you might have a five-year M&A attorney charging more than a thirty-year attorney who’s an associate. There’s some of that constraint still all within the same firm, within years of experience. It’s not always just what somebody will pay, which is so weird. But that’s how firms are.

Debbie: Yeah, no, I get it. And there’s some cultural barriers to overcome here, some tradition barriers to overcome here. There’s, yes, it’s a law firm, but for the love of it, it’s also a business.

Amanda: Yeah, I know, I know. You know me, I preach that on the daily.

Debbie: Yeah.

Rob: Okay, so

Debbie: Yeah.

Next week: strategic planning and budgeting season

Rob: So we’re already at time. I really wanted to get into what some of our members are asking for as well, but I’ll tease it a little bit. They want more, and they want a broader question base. Is that fair?

Amanda: I think we just went pretty broad today, so yeah.

Rob: I do, I think we should get into some of the strategic planning and budgeting season. Like what does AI mean? But that’ll have to wait for next week, because we are at time. I hope everybody has a great holiday weekend. Remember to join Sidebar, go to gosidebar.ai, the community’s becoming bigger and bigger, and you don’t want to miss out. We’ll see you next week.

Amanda: Bye all.